Stimulation of Human and Canine Neutrophil Metabolism by

Transcription

1 ANTIMICROBIAL AGENTS AND CHEMOTHERAPY, Feb. 1981, p /81/ $02.00/0 Vol. 19, No. 2 Stimulation of Human and Canine Neutrophil Metabolism by Amphotericin B SEE-RUERN SUPAPIDHAYAKUL,'2 LILIA R. KIZLAITIS,2 AND BURTON R. ANDERSEN' 2* Departments of Medicine and Microbiology, West Side Veterans Administration Medical Center,1 and University of Illinois Medical Center,2 Chicago, Illinois The effect of an antifungal agent, amphotericin B, on human and canine neutrophil metabolism was studied. Commercial preparations of amphotericin B in concentrations ranging from 5 to 100 yg/ml stimulated neutrophil chemiluminescence in the presence of 10- M luminol. This response was blocked by 2- deoxyglucose, a metabolic inhibitor, and by the absence of extracellular calcium ions. Neither pure amphotericin B nor the solubilizing agent present in the commercial preparation, alone or in combination, stimulated neutrophil chemiluminescence. Commercial amphotericin B caused an increase in oxygen uptake by neutrophils but no detectable superoxide anion production. Neutrophils were injured by commercial amphotericin B, as shown by an increase in trypan blue dye uptake but not cell lysis. Binding of amphotericin B to neutrophil membrane sterol with a subsequent alteration in membrane configuration is the most likely cause of metabolic stimulation. Amphotericin B, a polyene antibiotic, has been used in the treatment of systemic fungal disease since The major problem with its use has been toxicity (18); the intravenous form of amphotericin B frequently causes local thrombophlebitis, fever, chills, nausea, vomiting, headache, anorexia, hypokalemia, abnormal renal function, and anemia. These side effects limit the amount of drug that can be given intravenously. The antifungal action of amphotericin B depends upon its binding to cell membrane sterol (14) which causes potassium leakage and death of the fungus (15). These effects are blocked in the media by ergosterol, which binds to the amphotericin B and prevents it from adhering to the fungal cell membrane (16). Amphotericin B also can bind to human erythrocyte membranes, causing potassium leakage, free influx of sodium, and osmotic cell lysis (8). Free cholesterol can prevent erythrocyte lysis by competing with membrane cholesterol for amphotericin B binding (16). DeKruijef and Demel (12) suggested that eight amphotericin B-cholesterol complexes can combine on one side of a membrane to form half of a pore through the membrane lipid bilayer. A complete pore can result if a similar complex develops on the opposite membrane surface. Since the phospholipid:cholesterol ratio in neutrophil membranes is similar to that in erythrocyte membranes (20), it is likely that amphotericin B binds equally well to the neutrophil membrane. In 1976, Bjorksten et al. (6) studied the effects 284 of amphotericin B on neutrophil functions. Amphotericin B in a concentration of 2,ig/ml inhibited chemotactic responsiveness. Chemiluminescence of neutrophils as stimulated by opsonized zymosan was blocked by prior treatment of the cells with 5 ytg or more of amphotericin B per ml. These observations led us to study the amphotericin B-neutrophil interaction. MATERIALS AND METHODS Isolation of neutrophils. Neutrophils were prepared from healthy human volunteers or dogs. Blood anticoagulated with ethylenediaminetetraacetate (EDTA; 0.4 ml of 5% EDTA per 16 ml of blood) was separated by Ficoll-Hypaque gradient centrifugation (7) followed by dextran sedimentation (3 ml of 5% dextran per 10 ml of cell suspension; dextran molecular weight = 255,000) to allow the erythrocytes to sediment. Erythrocytes contaminating the neutrophil preparation were lysed with isotonic ammonium chloride solution (0.155 mm NH1C1, 10 mm KHCO1, 0.1 mm EDTA) at 4 C. The neutrophils were washed twice with calcium-magnesium-free Hanks balanced salt solution (HBSS) and then suspended in complete HBSS with the volume adjusted to provide the desired cell count. All neutrophil preparations contained 95 to 98% neutrophils. The cells were kept at 4 C before testing. Chemiluminescence measurement. Chemiluminescence was measured in a specially designed chemiluminescence spectrometer (1) which maintained the cells at 37 C with continuous mixing. The reaction volume was 5.0 ml per vial, and the neutrophil concentration was 106 cells per ml. Luminol (5-amino-2,3- dihydro-1,4-phthalazinedione, Eastman Kodak Co., Rochester, N.Y.; 108 M) was used to augment the

2 VOL. 19, 1981 light production, Emitted light was expressed as counts per minute, Background counts obtained from unstimulated neutrophils were subtracted from the values obtained during the experiment. The mean value for background counts was 5,570 cpm, with individual values ranging from 3,313 to 9,836 cpm in all studies. Oxygen uptake. Oxygen uptake was measured with an oxygen monitor and a Clark polarographic electrode (model 53; Yellow Springs Instrument Co., Yellow Springs, Ohio). Data are reported as microliters of oxygen consumed per 3 x 106 neutrophils in the first 12 min. Superoxide anion production. Superoxide anion was measured by the method of Babior et al. (3) with modification (13). The asay measures superoxide dismutase-inhibitable ferricytochrome c reduction by neutrophils. In the experiments to measure 02 uptake and superoxide anion production, opsonized zymosan was used as a positive control and a comparative stimulus. Zymosan (Sigma Chemical Co., St. Louis, Mo.), 4 mg/ ml in normal saline, was opsonized with 15% autologous serum at 37 C for 10 min. In all studies requiring opsonized zymosan, 0.11 mg was used per 106 neutrophils, a concentration which gives the optimal metabolic response. Trypan blue uptake and cell lysis. Trypan blue uptake was determined by incubating equal volumes of trypan blue dye (0.1%; GIBCO Laboratories, Grand Island, N.Y.) and neutrophils (106 cells per ml) for 15 min at 37 C. The percentage of cells containing trypan blue dye was determined by light microscopy. Neutrophil lysis after exposure to amphotericin B for 15 min at 370C was evaluated by counting the numbers of neutrophils and comparing this value with untreated cells (Coulter Counter, model Zf; Coulter Electronics, Inc., Hialeah, Fla.). Reagents. The commercial intravenous amphotericin B (Fungizone, E. R. Squibb & Sons., Inc., Princeton, N.J.) used in this study contained 50 mg of amphotericin B, 41 mg of sodium deoxycholate, 23 mg of Na2HPO4, and 2.2 mg of NaH2PO4 per vial. Ten milliliters of sterile water was used to reconstitute the amphotericin B, and 5% glucose solution was used for further dilution of the stock solution. The term "pure amphotericin B" designates a preparation without deoxycholate or phosphate (supplied by E. R. Squibb & Sons, Inc., New Brunswick, N.J.) which was dissolved in dimethyl sulfoxide in a concentration of 10 mg/ml and diluted further in 5% glucose adjusted to ph 9.0. The glucose solution at a higher ph improved the solubility of amphotericin B. Amphotericin B solutions were made up immediately before each experiment. Sodium deoxycholate (Fisher Scientific Company, Fair Lawn, N.J.), superoxide dismutase (Sigma Chemical Co.), 2-deoxyglucose (Sigma Chemical Co.), and N-ethylmaleimide (Mann Research Laboratories, Inc., New York, N.Y.) were reagent grade. STIMULATION OF NEUTROPHILS BY AMPHOTERICIN B 285 RES3ULTS Chemiluminescence. A commercial preparation of amphotericin B containing deoxycholate as a solubilizing agent and sodium phosphate as a buffer stimulated a chemiluminescent response in neutrophils when used in concentrations varying from 5 to 100 log/ml, but the highest and most reproducible chemiluminescence occurred at a concentration of 20,ig/ml. The peak of commercial amphotericin B-induced chemiluminescence had a day-to-day variation of 4.0 x 104 to 2.5 x i0' cpm in the presence of luminol (10-8 M). Without luminol we observed only a slight chemiluminescence. Human and canine neutrophils gave similar chemiluminescent responses. The light emission of neutrophils to pure amphotericin B (5 to 100jlg/ml) or sodium deoxycholate (4 to 42 jig/ml) was examined; also compared were mixtures of pure amphotericin B and sodium deoxycholate, with or without sodium phosphate buffer in the same proportions as found in the commercial preparation. Pure amphotericin B, sodium deoxycholate, and the mixtures gave only small chemiluminescent responses when compared with commercial amphotericin B (Fig. 1). Dimethyl sulfoxide alone in the concentration used in this experiment (less than 0.01%) did not stimulate or inhibit neutrophil chemiluminescence. Differences between the chemiluminescences of pure and commercial amphotericin B could not be explained by differences in solvents. When commercial amphotericin B was dissolved in the same manner as pure amphotericin B, it continued to induce a greater chemiluminescent response. The effect of 2-deoxyglucose, a known metabolic inhibitor of neutrophils (11), on the chemiluminescent response to amphotericin B was studied. Neutrophils were washed and resuspended in glucose-free HBSS. 2-Deoxyglucose (10 mm) was added to the cell preparation, and commercial amphotericin B was added as a stimulus. Commercial amphotericin B added to neutrophils in glucose-free HBSS served as a control. 2-Deoxyglucose caused a marked inhibition of neutrophil chemiluminescence (Fig. 2). When N-ethylmaleimide (0.1 mm), another metabolic inhibitor (11), was incubated with neutrophils, there was no chemiluminescent response to commercial amphotericin B as compared to a peak of 4.0 x 104 cpm in the absence of N-ethylmaleimide. The effects of divalent cations, calcium and magnesium, on chemiluminescence were studied. Neutrophils were resuspended in magnesium-free HBSS, calcium-free HBSS, or calcium- and magnesium-free HBSS. Neutrophils in complete HBSS (containing both 1.3 x 10-3 M calcium and 9.0 x 10-4 M magnesium) served as a control. Commercial amphotericin B was

3 286 SUPAPIDHAYAKUL, KIZLAITIS, AND ANDERSEN ANTIMICROB. AGENTS CHEMOTHER. 90,000 70, ,000 E C: 0. commercial amphotericin B U 30,000- amphote-icin B (!)Ure) desoxycholate amphoter-icin B (pure) _/z odesoxychlate phosphate 10,000 amhotric n B (poure ) desoxyhol ate ) T IME(min) FIG. 1. Chemiluminescence of canine neutrophils (5 x 106 cells) to pure amphotericin B (20 pg/ml), commercial amphotericin B (20 plg/ml), sodium deoxycholate (16.4 Ag/ml), and pure amphotericin B (20,ug/ ml) plus sodium deoxycholate (16.4 pg/ml) with and without phosphate buffer. This experiment was performed in the presence of luminol (10-' M). added, and the chemiluminescent response of neutrophils was observed. In the presence of calcium even without magnesium the chemiluminescent response was not significantly different from the control; but in the absence of calcium even in the presence of magnesium, there was no significant chemiluminescence (Fig. 3). Oxygen uptake studies. Commercial amphotericin B (20 gg/ml) caused an increase in oxygen uptake of only 0.48 [LI of oxygen per 3 x 106 human neutrophils in the first 12 min (range, 0.27 to 0.86 [li) compared with opsonized zymosan, which induced an oxygen uptake of 2.57 ul of oxygen per 3 x 106 neutrophils in the first 12 min. Superoxide anion assay. There was no detectable superoxide anion production by 3 x 10" canine neutrophils in the presence of 20,ig of commercial amphotericin B per ml as compared with opsonized zymosan, which stimulated neutrophils to reduce 88 nmol of superoxide-dependent cytochrome c per 3.0 x 106 neutrophils (range, 60 to 120 nmol). Trypan blue uptake and cell lysis. Commercial amphotericin B ranging from 5 to 100 tig/ml did not cause lysis of canine neutrophils observed over 15 min at 37 C when compared with control cells. Increased trypan blue uptake was noted as the concentration of commercial amphotericin B increased. At a concentration of 20 [ig of commercial amphotericin B per ml, 17c

4 VOL.- 19, 1981 STIMULATION OF NEUTROPHILS BY AMPHOTERICIN B 287 Q)! c E w 0- Y-1 c :3 8 2 IC -A--u TI ME (min) FIG. 2. Chemiluminescence of human neutrophils (5 x 106 cells) to commercial amphotericin B (20 pg/ ml), in glucose-free HBSS with (dashed line) and without (solid line) 2-deoxyglucose (10 mm), in the presence of 108 M luminol. Brackets indicate one standard deviation. of the neutrophils were nonviable as determined by the inability to exclude the dye, compared with 2% of control neutrophils. DISCUSSION Studies by Bjorksten et al. (6) have shown that amphotericin B (5 to 20,Ig/ml) pretreatment of neutrophils suppressed the chemiluminescent response to opsonized zymosan. These authors, however, did not examine the capacity of amphotericin B itself to stimulate neutrophils to emit light. Our study has shown that commercial amphotericin B induced strong chemiluminescence in the presence of luminol, whereas pure amphotericin B plus luminol gave a much weaker response. The chemiluminescence of commercial amphotericin B was not due to sodium deoxycholate used as a solubilizing agent in the commercial preparation because there was no light emission from neutrophils in contact with sodium deoxycholate alone. When the mixture of pure amphotericin B and sodium deoxycholate with or without phosphate was used to stimulate neutrophils, the light emission was still much lower than from commercial amphotericin B. Commercial amphotericin B is a submicroscopic colloidal suspension (4), whereas pure amphotericin B and other mixtures used in our study were not colloids. In the colloidal form commercial amphotericin B appears to have a greater in vivo effect and increased toxicity to both animals and humans (5). Our studies indicated that the colloidal state of amphotericin B as found in the commercial preparation was required for optimal stimulation of neutrophil chemiluminescence. In 1977, Cohen and Chovaniec (11) showed that the metabolic response of neutrophils harvested from the peritoneal cavity of guinea pigs was blocked by 2-deoxyglucose and N-ethylmaleimide. This was determined by measuring the production of superoxide anion. In the present study the chemiluminescent response to commercial amphotericin B was markedly inhibited by 2-deoxyglucose and N-ethylmaleimide, which suggests that metabolic stimulation was responsible for the chemiluminescence induced by commercial amphotericin B. However, as compared with opsonized zymosan, commercial amphotericin B caused only a small amount of oxygen uptake in neutrophils, and no superoxide anion could be detected. This suggests that the chemiluminescence from amphotericin B- treated neutrophils is the result of a low-level metabolic response that results in some oxygen uptake but too little superoxide production to be detectable. The mechanism by which amphotericin B caused stimulation of neutrophil metabolism and chemiluminescence probably involves the binding of amphotericin B to cholesterol on neutrophil membranes. DeKruijef and Demel (12) have shown that amphotericin B can bind to cholesterol to form complexes in the membranes of Acholeplasma laidawii cells, causing pores through half the thickness of the membrane. The hydrophilic channel of this pore has a diameter of about 0.8 nm. Van Zutphen et al. (19) and Cass et al. (9) found that this half-pore decreased the membrane electrical resistance to a significant extent, and Cass et al. (9) found that it is anion selective and discriminates among anions on the basis of their size. Previous data (8, 9, 12, 15, 19) have shown that amphotericin B can cause cell membrane injury. Our data demonstrated an increase in trypan blue dye uptake by neutrophils after incubation with amphotericin B, suggesting that it was causing an increase in membrane permeability to the dye. Chunn et al. (10) have shown that the neutrophil toxicity of amphotericin B as measured by trypan blue uptake could be blocked by cholesterol, suggesting that the free cholesterol was competing with membrane cholesterol binding sites of amphotericin B. This amphotericin B-cholesterol interaction may be the primary event leading to neutrophil metabolic stimulation. The activation of neutrophil metabolism secondary to binding with membrane cholesterol is analogous to the stimulatory effect of streptolysin 0 on neutrophils as reported by Andersen and Duncan (2). Streptolysin 0, which has no known enzymatic activity, appears to be

5 288 SUPAPIDHAYAKUL, KIZLAITIS, AND ANDERSEN 120,000 ANTIMICROB. AGENTS CHEMOTHER. Ca+ (1.3 x 10 3M) M) a) -4- c E... 60,00( cn a-) c 0. :3 / / Mg (9.0 x 10 4M) Ca ++ + Mg + FRE E r..=-1l..4i * I *l I I I I I T I ME (min) FIG. 3. Chemiluminescence of human neutrophils (5 x 10' cells) to commercial amphotericin B (20 pg/ml) in HBSS (Ca2` = 1.3 x 10-3 M; Mg2+ = 9.0 x 10-4 M), and in the absence of Ca2` or Mg2+ or both. This experiment was performed in the presence of luminol (10' M). Brackets indicate one standard deviation. capable of stimulating metabolic activity by virtue of its cholesterol-binding ability. The requirement of extracellular calcium for the neutrophil metabolic response to the stimulus formylmethionylleucylphenylalanine has been shown by Simchowitz and Spilberg (17). We also showed that extracellular calcium was necessary in the chemiluminescent response of neutrophil to commercial amphotericin B. Amphotericin B appears to have caused some membrane alteration which results in a calcium-dependent stimulation of neutrophil metabolism. These studies indicate that amphotericin B in the colloidal state can cause metabolic activation of neutrophils and an associated chemiluminescence. The toxic properties of amphotericin B may be, in part, related to this phenomenon. ACKNOWLEDGMENTS We thank Rosalind R. Parker for her excellent secretarial assistance. This work was supported by West Side Veterans Administration Hospital research funds (MRIS 0394). LITERATURE CITED 1. Andersen, B. R., and A. M. Brendzel Use of a unique chemiluminescence spectrometer in a study of factors influencing granulocyte light emission. J. Immunol. Methods 19: Andersen, B. R., and J. L. Duncan Activation of human neutrophil metabolism by streptolysin 0. J. Infect. Dis. 141: Babior, B. M., R. S. Kipnes, and J. T. Curnutte Biological defense mechanisms: the production by leukocytes of superoxide, a potential bactericidal agent. J. Clin. Invest. 52: Bartner, E., H. Zinnes, R. A. Moe, and J. S. Kuleaza Studies on a new solubilized preparation of amphotericin B. Antibiot. Annu Bennett, J. E., G. J. Hill II, W. T. Butler, and C. W. Emmons Correlation of particle size of intravenous amphotericin B with toxic and chemotherapeutic effects. Antimicrob. Agents Chemother. 3: Bjorksten, B., C. Ray, and P. G. Quie Inhibition of human neutrophil chemotaxis and chemiluminescence by amphotericin B. Infect. Immun. 14: Boyum, A Isolation of mononuclear cell and granulocytes from human blood: isolation of mononuclear cells by one centrifugation and of granulocytes by combining centrifugation and sedimentation at Ig. Scand. J. Clin. Lab. Invest. Suppl. 97: Butler, W. T., D. W. Alling, and E. Cotlove Potassium loss from human erythrocytes exposed to amphotericin B (29825). Proc. Soc. Exp. Biol. Med. 118: Cass, A., A. Finkelstein, and V. Krespi The ion permeability induced in thin lipid membranes bv the polyene antibiotics nystatin and amphotericin B..J.

6 VOL. 19, 1981 STIMULATION OF NEUTROPHILS BY AMPHOTERICIN B 289 Gen. Physiol. 56: Chunn, C. J., P. R. Starr, and D. N. Gilbert Neutrophil toxicity of amphotericin B. Antimicrob. Agents Chemother. 12: Cohen, H. J., and M. E. Chovaniec Superoxide generation by digitonin-stimulated guinea pig granulocytes. J. Clin. Invest. 61: DeKruijef, B., and R. A. Demel Polyene antibiotic-sterol interactions in membrane of Acholeplasma laidlawii cells and lecithin liposomes. Biochim. Biophys. Acta 339: Harvath, L., H. J. Amirault, and B. R. Andersen Chemiluminescence of human and canine polymorphonuclear leukocytes in the absence of phagocytosis. J. Clin. Invest. 61: Kinsky, S. C Membrane sterols and the selective toxicity of polyene antifungal antibiotics. Antimicrob. Agents Chemother. 3: Kotler-Brajtburg, J., G. Medoff, G. S. Kobayashi, S. Boggs, D. Schlessinger, R. C. Pandey, and K. L Rinehart, Jr Classification of polyene antibiotics according to chemical structure and biological effects. Antimicrob. Agents Chemother. 15: Kotler-Brajtburg, J., H. D. Price, G. Medoff, D. Schlessinger, and G. S. Kobayashi Molecular basis for the selective toxicity of amphotericin B for yeast and filipin for animal cells. Antimicrob. Agents Chemother. 5: Simchowitz, L., and I. Spilberg Generation of superoxide radicals by human peripheral neutrophils activated by chemotactic factor. Evidence for the role of calcium. J. Lab. Clin. Med. 93: Utz, J. P., J. E. Bennett, M. W. Brandriss, W. T. Butler, and G. J. Hill I Amphotericin B toxicity. Ann. Intern. Med. 61: Van Zutphen, H., R. A. Demel, A. W. Norman, and L L. M. Van Deenen The action of polyene antibiotics on lipid bilayer membrane in the presence of several cations and anions. Biochim. Biophys. Acta 241: Woodin, A. M., and A. A. Wieneke Composition and properties of a cell-membrane fraction from polymorphonuclear leucocyte. Biochem. J. 99: Downloaded from on November 13, 2018 by guest